iterating-with-ai-and-mcp

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill defines tools such as hotswan_reload and hotswan_build_and_install to perform build and deployment tasks. These tools are executed within the development environment to facilitate the hot-reload lifecycle.
  • [INDIRECT_PROMPT_INJECTION]: The agent is instructed to ingest and analyze external data, including Kotlin source files and device screenshots (hotswan_take_screenshot). This establishes a surface where the agent could potentially encounter malicious content in the application under test. The skill structure mitigates this by using a specific verification loop and requiring status checks before execution.
  • [EXTERNAL_DOWNLOADS]: The skill links to official documentation and plugin listings on GitHub and the JetBrains Marketplace. These references are used for configuration and guidance purposes and point to the official resources for the Compose HotSwan toolset.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 06:53 AM
Security Audit — agent-trust-hub — iterating-with-ai-and-mcp